Spectral time domain induced polarization : factors affecting spectral data information content and applicability to geological characterization

anglais. The geo-electrical survey (resistivity and induced polarization) with different settings, arrangements, time base, spreads and waveforms (50% and 100% duty cycle waveform) was carried out over the Liikavaara depo-sit, near the Aitik mine, in northern Sweden and all the measured data was inverted in two different softwares: Res2Dinv and AarhusInv. Surveys of induced polarization (IP) often contain data of low quality, due to the high sensitivity against noise sources and coupling effects. Measured data with the new method (applying 100% duty cycle waveform), which is much faster than the conventional one (50% duty cycle waveform), present significant improvements in the quality of IP-data. It was also attempted to increase the data quality by separating current and potential cables and restrict the capacitive coupling, which normally occurs between transmitting current and pot-ential receiving wires. Moreover, conceptual models from different measurements were correlated to boreholes close to the survey line, in order to find geological correlation between the models and logging data. There was also a possibility of modeling Tau and C parameters by the newly developed software (AarhusInv) which gave us more detailed information about the mineralization texture (grain size and grain size distribution). They were also compa-red in different measurements and it appears that Tau and C are sensitive to the duration of current injection. However, lack of reference data for interpretation, made the interpretation of these two parameters, difficult and ambiguous.

anglais. Since the 1900s electrical resistivity and induced polarization methods are one of the common geophysical survey techniques which are applied in subsurface studies and in particular in mineral exploration. In this paper we show how different settings, arrays, waveforms, cable spreads and time base can affect the data quality/ the final inversion models for resistivity and IP . The results from this study show that the acquired IP data are of higher quality when applying 100 % duty cycle waveforms (measuring IP during on time), and an additional advantage is decreasing the measuring time. It was also attempted to increase the data quality by separating current and potential cables and restrict the capacitive coupling, which normally occurs between transmitting current and pot-ential receiving wires and as result separated spreads data showed higher data quality (fewer number of noisy curves), particularly for the early time windows. The final models were also compared and it appears that different injection time does not show any effect on the final models in resistivity and IP. However, spectral IP parameters (Tau and C) appear to be sensitive to the duration of current injections in which; decreasing the time base causes diminished anomalies in Tau and partly in C. Measurement with single or separated spreads does not have any significant effect on these parameters. Moreover correlation of sulfide content matched well with the inverted models, which confirms the applicability of the DCIP method in mineral exploration.